Discrete-time Temporal Network Embedding via Implicit Hierarchical Learning in Hyperbolic Space

Yang, Menglin; Zhou, Min; Kalander, Marcus; Huang, Zengfeng; King, Irwin

doi:10.1145/3447548.3467422

Cited by 62 publications

(32 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note that the proposed idea is decoupled from the collaborative filtering, and theoretically, they can be applied to other hyperbolic models. In future work, we will extend our idea to more recommendation scenes or non-recommendation scenarios, like the hyperbolic temporal link prediction [46].…”

Section: Discussionmentioning

confidence: 99%

HRCF: Enhancing Collaborative Filtering via Hyperbolic Geometric Regularization

Yang

Zhou

Liu

et al. 2022

Proceedings of the ACM Web Conference 2022

Self Cite

View full text Add to dashboard Cite

In large-scale recommender systems, the user-item networks are generally scale-free or expand exponentially. The latent features (also known as embeddings) used to describe the user and item are determined by how well the embedding space fits the data distribution. Hyperbolic space offers a spacious room to learn embeddings with its negative curvature and metric properties, which can well fit data with tree-like structures. Recently, several hyperbolic approaches have been proposed to learn high-quality representations for the users and items. However, most of them concentrate on developing the hyperbolic similitude by designing appropriate projection operations, whereas many advantageous and exciting geometric properties of hyperbolic space have not been explicitly explored. For example, one of the most notable properties of hyperbolic space is that its capacity space increases exponentially with the radius, which indicates the area far away from the hyperbolic origin is much more embeddable. Regarding the geometric properties of hyperbolic space, we bring up a Hyperbolic Regularization powered Collaborative Filtering (HRCF) and design a geometric-aware hyperbolic regularizer. Specifically, the proposal boosts optimization procedure via the root alignment and origin-aware penalty, which is simple yet impressively effective. Through theoretical analysis, we further show that our proposal is able to tackle the over-smoothing problem caused by hyperbolic aggregation and also brings the models a better discriminative ability. We conduct extensive empirical analysis, comparing our proposal against a large set of baselines on several public benchmarks. The empirical results show that our approach achieves highly competitive performance and surpasses both the leading Euclidean and hyperbolic baselines by considerable margins. Further analysis verifies the rationality and effectiveness of the proposal for robust, deeper, and lightweight neural graph collaborative filtering.

show abstract

Section: Discussionmentioning

confidence: 99%

HRCF: Enhancing Collaborative Filtering via Hyperbolic Geometric Regularization

Yang

Zhou

Liu

et al. 2022

Proceedings of the ACM Web Conference 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…One common way to handle temporal graphs is to decompose it into multiple static graphs snapshots by a regular time interval. Some works embed the graph convolution into the recurrent neural network (RNN) based models or attention mechanism [33], which learns to exploit the dynamic information in the graph evolution within a period of time [13], [14], [16], [17], [34], [35], [36]. Some other works are dynamic extensions of ideas applied in the static case inspired by methods such as PageRank [37], graph autoencoder [38] and the topic model [15] to capture both the temporal community dynamics and evolution of graph structure.…”

Section: Temporal Graph Embeddingmentioning

confidence: 99%

ConTIG: Continuous Representation Learning on Temporal Interaction Graphs

Xu¹,

Fan²,

Yang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Representation learning on temporal interaction graphs (TIG) is to model complex networks with the dynamic evolution of interactions arising in a broad spectrum of problems. Existing dynamic embedding methods on TIG discretely update node embeddings merely when an interaction occurs. They fail to capture the continuous dynamic evolution of embedding trajectories of nodes. In this paper, we propose a two-module framework named ConTIG, a continuous representation method that captures the continuous dynamic evolution of node embedding trajectories. With two essential modules, our model exploit three-fold factors in dynamic networks which include latest interaction, neighbor features and inherent characteristics. In the first update module, we employ a continuous inference block to learn the nodes' state trajectories by learning from time-adjacent interaction patterns between node pairs using ordinary differential equations. In the second transform module, we introduce a self-attention mechanism to predict future node embeddings by aggregating historical temporal interaction information. Experiments results demonstrate the superiority of ConTIG on temporal link prediction, temporal node recommendation and dynamic node classification tasks compared with a range of state-of-the-art baselines, especially for long-interval interactions prediction.

show abstract

“…Link prediction aims to predict whether two nodes in a network are likely to have a link [15,27,34] via the partially available topology or/and node attributes, attracting considerable research efforts owning to its diverse applications, such as the friend recommendation in social networks [1,30], the item recommendation [12,26,36] in user-item networks, location recommendation in urban computing [13,14], the relation completion in knowledge graphs [18,43],…”

Section: Introductionmentioning

confidence: 99%

“…drug-target prediction in biomedical networks [17,20,42], and disease infectious analysis of epidemic spread [4,7,34].…”

Section: Introductionmentioning

confidence: 99%

BSAL: A Framework of Bi-component Structure and Attribute Learning for Link Prediction

Li,

Zhou,

Zhang

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Given the ubiquitous existence of graph-structured data, learning the representations of nodes for the downstream tasks ranging from node classification, link prediction to graph classification is of crucial importance. Regarding missing link inference of diverse networks, we revisit the link prediction techniques and identify the importance of both the structural and attribute information. However, the available techniques either heavily count on the network topology which is spurious in practice, or cannot integrate graph topology and features properly. To bridge the gap, we propose a bicomponent structural and attribute learning framework (BSAL) that is designed to adaptively leverage information from topology and feature spaces. Specifically, BSAL constructs a semantic topology via the node attributes and then gets the embeddings regarding the semantic view, which provides a flexible and easy-to-implement solution to adaptively incorporate the information carried by the node attributes. Then the semantic embedding together with topology embedding are fused together using attention mechanism for the final prediction. Extensive experiments show the superior performance of our proposal and it significantly outperforms baselines on diverse research benchmarks.

show abstract

Discrete-time Temporal Network Embedding via Implicit Hierarchical Learning in Hyperbolic Space

Cited by 62 publications

References 23 publications

HRCF: Enhancing Collaborative Filtering via Hyperbolic Geometric Regularization

HRCF: Enhancing Collaborative Filtering via Hyperbolic Geometric Regularization

ConTIG: Continuous Representation Learning on Temporal Interaction Graphs

BSAL: A Framework of Bi-component Structure and Attribute Learning for Link Prediction

Contact Info

Product

Resources

About